Main Article Content

Quality of Health Data in Public Health Facilities of Oromia and Gambela Regions, Ethiopia


Muluemebet Abera Wordofa
Feyissa Tolessa Garedew
Gurmessa Tura Debelew
Berhane Megerssa Ereso
Dawit Wolde Daka
Gelila Abraham
Tilahun Fufa Debela
Yisalemush Asefa Demissie
Ketema Lemma Abdi
Meskerem Seboka Ergiba
Asaye Birhanu Senay
Hailu Dawo Mio
Solomon Kassahun Gelaw
Chaine Hussen Ibrahim
Tamiru Regassa Turi
Lamessa Tadesse Amante
Abinet Feyisa Aredo
Negalign Berhanu Bayou

Abstract

Background: The quality of routine health data in Ethiopia remains poor despite the existence of proven effective improvement strategies. Implementation research frameworks assist in analyzing, operationalize, and assess the implementation process by systematically addressing the "know-do" gap. The study aimed to assess routine health data quality status in public health facilities of selected districts in Oromia and Gambella regions of Ethiopia using the Consolidated Framework for Implementation Research (CFIR).
Methods: This study is part of a larger implementation research project that employed an institutional-based cross-sectional study design. A pre-intervention baseline assessment was conducted using qualitative and quantitative data collection methods. All health centers and hospitals in the selected districts and selected health posts based on their performance, one high, one medium and one low performer, a total of three from each primary health care unit, were included in the study. As a result, two hospitals, seven health centers, 35 health posts, and both district health offices were included in the study. Moreover, a total of 51 key informant interviews, four participants from each health facility, who have more connection with HMIS data like facility head, HMIS focal, MCH and outpatient department heads as well as HMIS focal person from the district, District health office head, plan and MCH heads one health extension worker per selected health post were included purposely. Structured checklists were developed from the Performance of Routine information system management (PRISM) assessment tool, which is nationally approved and used for quantitative data. For the qualitative data, semi-structured interview guides were developed by reviewing different literature. Descriptive analysis was done for quantitative data, while a thematic analysis approach was used for qualitative data.
Results: A total of 46 HFs in 2 districts were involved in the study. Data quality in terms of accuracy, timeliness and completeness was assessed for eight selected indicators. Accordingly, three of the assessed data elements were in the acceptable range. These were delivery service (98%), Penta-3 (96%) and Measles (94%). The remaining 5 were out of the acceptable range, indicating the presence of over-reporting. In both districts, report completeness was 100% regarding timeliness 53.5% of the report in Digalu-Tijo and 79% in Godere submitted to the next level according to the national schedule for each level. Data were inconsistent both over time and between indicators. Supportive supervision and higher-level mentorship support were inadequate, although the extent varies between districts. The gaps were more pronounced in the Digalo-Tijo district than in the others. Internal supervision and mentorship were also missing in both contexts. Data quality review meetings were not conducted regularly. Many recording and reporting tools were unavailable on the assessment day or never available in health facilities. Many health workers did not receive training during the last 12 months on health management information system-related topics.
Conclusion: Data generated through routine health management information systems is generally low quality. Supportive supervision, mentorship, and review meetings are not accomplished as indicated in the strategy. Many health facilities lack important recording and reporting tools. Training on health management information system-related topics is inadequate. The findings highlight the need to design tailored and context-specific interventions to improve data quality.


Journal Identifiers


eISSN: 1021-6790